Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: probable Kasparov ting.

Author: Robert Hyatt

Date: 06:39:24 01/10/00

Go up one level in this thread


On January 10, 2000 at 02:15:09, Ed Schröder wrote:

>On January 09, 2000 at 23:13:58, Robert Hyatt wrote:
>
>>On January 09, 2000 at 21:10:29, Mark Young wrote:
>>
>>>On January 09, 2000 at 18:33:26, Enrique Irazoqui wrote:
>>>
>>>>On January 09, 2000 at 17:43:32, Robert Hyatt wrote:
>>>>
>>>>>Let's not go off the deep end here.  I personally don't give a care about
>>>>>his comments.  Feel free to tell him to drop by ICC at any point in time he
>>>>>chooses and demonstrate how badly Crafty plays in a few blitz games.  He
>>>>>reminds me of the "catfish" in the deep south waterways.  Eat anything.  Dumb
>>>>>as a rock.  We say "he is like a catfish, all mouth and no brains" when some-
>>>>>one makes silly and easily refutable remarks...
>>>>>
>>>>>Kasparov is in ChessBase's pocket.  However, from this point forward, since
>>>>>we are going to continue to see this, I believe that I will simply choose to
>>>>>say that "no more crafty versions will be available for ChessBase products".
>>>>>
>>>>>If it is so bad, they really don't need it anyway.  If I see any future versions
>>>>>on their web site, I will let my attorney do the talking.
>>>>
>>>>The article is poisonous and I understand your anger, but I think you are not
>>>>aiming well. First of all, I can't imagine Kasparov being in anybody's pocket.
>>>>Second, the heated argument about Deep Blue can easily explain the animosity
>>>>towards Crafty without resorting to third parties.
>>>>
>>>>What I can tell you is that I keep receiving regular updates of Crafty for CB,
>>>>together with positive comments about it. I don't believe in the long and
>>>>sinister arm of CB, and even less in this case.
>>>>
>>>>Enrique
>>>
>>>I don't see why Bob is so upset. The comments made are pretty much accurate.
>>
>>Actually they are way off, and here is why:
>>
>>
>>>
>>>"Crafty is extremely strong in tactics." ....True
>>
>>Dead false. Try _any_ commercial program on a tactical test position, and then
>>try Crafty.  I have not tried to fiddle a lot with deep search extensions, and
>>am usually far slower than everybody else to find a deep tactical shot.  The
>>King may be the best, but others like Rebel are very good.  I do _not_ consider
>>crafty very tactical.
>
>I remember Crafty solves the WAC position, key-move Rb4! Quite impressive!
>

Cray Blitz solved it in a couple of seconds.  _that_ is tactical.  Crafty takes
over a minute on the quad xeon (finds it at depth 16).  I don't consider that
nearly so tactical.

But that is only one kind of tactics (non-king-attack).  In king attack type
positions, I don't do very well speed-wise, but don't really spend a lot of
time worrying about this yet.




>
>
>>>"it makes positional errors and misplays some endgames" ....True
>>
>>It misplays fewer endgames than most commercial programs.  It certainly has
>>some knowledge that none of them at present are using.
>>
>>
>>
>>
>>>
>>>"It doesn?t understand piece mobility." ....True, but could use more
>>>explaination.
>>
>>Dead false.  I use mobility where it is important, because my bitmap stuff
>>makes it totally free to do so.  IE it pretty well understands good vs bad
>>bishops, due to this.  etc.
>>
>>
>>
>>>
>>>"Crafty is a weak program compared with the best commercial chess
>>>programs."....True
>>>
>>>I have seen no crafty results even by you that dispute those observations. Bob
>>>may not like it but at this time his program is still weaker then the top
>>>programs. It must be noted that bob has closed the gap in the last year, but
>>>crafty is still not on par with the top commercail programs.
>>
>>
>>I play the top commercial programs on ICC all the time, at long standard
>>time controls.  So far, nobody is 'killing me'.  I can't think of a program
>>that I am not at least breaking even with, and really can not think of one
>>of those.  Yes I have faster hardware.  But I spent the time to make the program
>>use it...
>>
>>But the real fly in the ointment was to say it is not playing well, then give
>>a tournament where it finished in second place above the programs the guy was
>>praising...  and in that tournament it was on _equal_ hardware.
>>
>>You might notice one thing in my writing here and elsewhere.  I don't run down
>>other programs.  Because I _know_ that every chess program represents a heck of
>>a lot of work.  De Gorter obviously doesn't know much about it at all...  and >it shows...
>
>It's just the opinion of one man. And Crafty has proven itself since
>times, no?
>
>Ed


Yes.. but it is the sort of thing that should not be written about _any_
program.  Much less a freeware program in the context of a review of
"commercial" chess engines.




This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.